The Pakistan Development Review

42 : 4 Part II (Winter 2003) pp. 727750

The Relationship between Economic Growth and

Capital Structure of Listed Companies: Evidence
of Japan, Malaysia, and Pakistan
MUHAMMAD MAHMUD*
1. INTRODUCTION
Corporate enterprise is a natural outcome of capitalism in the course of
economic development. The underwriter firms and banks etc. initially meet the
capital requirements of such enterprise. Later on it is the stock exchange that carries
out redistribution of shares of the enterprise.
Corporate decisions on capital structure policy have long been a subject of debate
and still remain an unresolved issue. The traditional view of capital structure was that it
results in the weighted average cost of capital being U-shaped, which means that there
exists as an optimal mix between debt and equity, at which point a firms value is
maximised. However, Modigliani-Miller (1958), in a world of no tax and no financial
distress, proved that capital structure is irrelevant to explaining firm values. When
company taxes are considered, the benefits from tax shield leads Modigliani-Miller
(1963) to conclude that the value maximising capital structure is extreme leverage. In a
subsequent paper Miller (1977), by introducing both corporate tax and personal taxes into
the model, points towards irrelevance of capital structure for any particular firm.
Firms behaviour in reality seems to indicate that it conforms to some
acceptable mix between debt and equity. Behaviourally this can be explained by
the tradeoff theory. The theory says that there are two opposing forces at work for a
levered company. The positive forces are derived from tax savings due to the
creation of interest tax-shield and the ensuing management discipline when
companies employ debt in their capital structure. The negative forces are those
associated with over leveraging, a situation where risk of default is reasonably high.
The tradeoff theory is consistent with the traditional view that there exist an optimal
mix between debt and equity.
Capital market is an important supplier of funds for economic and industrial
growth of the country. Investment in infrastructure is necessary for a strong, flexible,
and growing economy. Various country economic factors are analysed. However, the
Muhammad Mahmud is Assistant Professor, Institute of Business Administration, Karachi.
Authors Note: I am thankful to Dr Abdul Qayyum of the Pakistan Institute of Development
Economics, Islamabad, for his useful comments and suggestions on an earlier draft of this paper.
However, any errors are solely the responsibility of the author.

Muhammad Mahmud

728

relationship between market capitalisation and GDP growth is not linear. At a certain
level, the tax burden associated with financing and maintaining public capital
reduces the returns to private industry, which, in turn, reduces growth.
2. OBJECTIVES OF THE STUDY
The current study is an attempt to add to existing literature on the relationship
between economic growth and capital structure decisions of firms in three Asian
countries: Japan, Malaysia and Pakistan. These countries are chosen in order to
represent three different stages of economic development. One can hypothesise that
capital market develops in tandem with general economic development. As capital
market develops, firms tend to use more debt financing, as evidenced from various
other studies [see for example Rajan and Zingales (1995)].
The specific objectives are as follows:
(1) to investigate if countrys economic factors play a significant role in
determining capital structure between markets;
(2) to investigate if capital structure is different across industry class in each
country and across countries;
(3) to investigate firm specific factors influencing capital structure decision in
each of the three countries. These factors are growth, size, fixed asset ratio,
profitability, operating leverage and dividend policy;
(4) to examine the variations in the leverage ratios across countries, and as to
whether such variations depend on macroeconomic variables.
3. PREVIOUS STUDIES
Author(s)
Toy, et al. (1974)

Scope of Study
1966-1972, France, Japan,
Holland, Norway and U.S.

Remmers, et al. (1974)

Nakamura and Nakamura

(1982)

1967-1972, Fortune 500

companies
1967-1972, 12 industries
1969-1976, 233 non-regulated
firms
1964-1974 for US firms
1966-1970 for Japan firms

Titman and Wessels

(1983, 1988)

1974-1982, 469
manufacturing firms

Scott and Martin (1975)

Ferry and Jones (1979)

Results
Debt ratios, asset growth (+),
earnings variability (+),
earnings rate ()
Industry (0), size (0)
Industry class (s), size (s)
Industry (s), size (), earnings
variability (0)
Retained earnings (), cost of
debt (), capital productivity(),
cost of equity (+)
Profitability (), size (),
earnings variability (0),
collateral value of fixed asset
(0), future growth (0), non-debt
tax shield (0), industry class
(yes), asset uniqueness ()

Continued

Economic Growth and Capital Structure of Companies

Author(s)
Baskin (1989)

Scope of Study
1960-1972, 378 Fortune
500 firms
1969-1987, 508 US firms

Chang and Rhee (1990)

Graham and Bromson (1992)

1980-1989, Australia,
Hong Kong, Japan, South
Korea, Taiwan and Thailand

Homaifer, et al. (1994)

Rajan and Zingales (1995)

1978-1988, 370 US firms

Japan, Germany, France, Italy,
UK, Canada and US.

Johnson (1997)
1985-1989, 847 US firms
Note: + means positively related to leverage.
means negatively related.
0 means no significant relationship.
s means significant.
? means mixed results.

729

Results
Growth (+), profits (),
dividend (+)
Profitability (), non-debt tax
shield (+), growth (+), size (+),
earnings variability (+)
Country effect (s), industry
(Yes for Japan, S. Korea, No
for Australia, Hong Kong,
Taiwan,
Thailand)
Size (+) and future growth (+)
Accounting regulation (s),
institutional environment (s),
tangibility (+), market to book
ratio (), logsale (+),
profitability ()
Fixed asset ratio (+)

4. DATA AND METHODOLOGY

This study is the updated sample data that contained in my PhD. research
thesis (2000). The data was derived from various sources. Firm specific variables
are obtained from company annual reports. The annual reports and macroeconomic
variables are obtained from the Extel Company Research Services and Daiwa
Securities Research Institute for Japan; PACAP Database of the University of Rhode
Island for Malaysia; and Vital Information Services, Karachi for Pakistan. The
sample includes only non-financial companies. It is also required that the companies
are listed on the respective stock exchanges of the three countries over the ten-year
period covered in this study, 1989 until 1998. The sample consists of 505 (29 percent
of the number of companies on the stock exchange in 1998) firms from Japan, 109
(30 percent of the number of main board companies) from Malaysia and 104 (24
percent) from Pakistan.
There are several commonly used debt ratios in studies on capital structure. In
the context of this study, our main issue of investigation is laid out on the premise of
the static tradeoff theory, which, in simple terms states that some amount of debt is
desirable, but too much of it brings in financial distress. In this light we are
concerned with the total amount of debt used by a firm to finance its entire operation
and firms ability to service the loans. We should therefore be concerned with total
debt and total liability of the firms. The distinction between short-term and long-term
debt is less useful due to the ability of firms to rollover short-term loans.

730

Muhammad Mahmud

Nevertheless it would also be interesting and useful to study the behaviour of

long-term debt because it traditionally forms an important component of capital.
In this study three leverage measures are used.
(1) Total liability (non-equity) to total asset ratio (TLA). The liability includes
short-term liabilities and long-term loans. This measure reflects the
amount of claims in book value terms in case of liquidation of the firm. It
also reflects the residual value to shareholders.
(2) Total debt to equity (TDE). Total debt includes short-term loan and
overdrafts, and long-term liabilities. This is truly a measure of leverage in
the sense that fixed interest commitment acts as a lever to enlarge return to
shareholders. This ratio is a linear transformation of another commonly
used leverage measure, total debt to capital.
(3) Long-term debt to capital. Capital in this measure is defined as longterm-debt plus book value of equity. This truly measures the long-term
financing mix of the company.
Based on earlier empirical studies on capital structure, it is found that there is
quite a long list of firm-specific variables that have been investigated. In this study we
tried to include as many variables as permitted by the data in order to have a
comprehensive study on relationship between economic growth and capital structure
in the three countries. The variables finally included are checked to ensure they are not
highly correlated with one another. The following variables are included on our study:
(1) GRA. Growth in assets is defined as compounded annual growth of book
value of total assets beginning from year 1989 until 1998. If growth in
assets is high, firms may have to resort to external funding to finance its
operation. In other words, if firms are growing at a faster pace than can be
finance by internally generated funds, they have to resort to external
funding. Depending on the pecking order of financing, growth may be
expected to have a positive or a negative relationship with debt ratios. If
firms prefer debt financing to new equity as in the US situation, positive
relationship may be expected. But in Malaysia and many other Asian
countries firms prefer new equity to debt [see for example, Kester and Isa
(1994)]. In this case high growth may not be associated with high debt
ratio.
(2) GRS. Growth in sales is defined as compounded annual growth rate of
sales beginning from year 1989 until 1998. The expected relationship is
similar to that for Growth in Assets (GRA). However, firms with high
sales growth may experience a rapid increase in trade receivables and
payables, and this may result in high total liabilities ratio but not
necessarily increasing the debt ratios.

Economic Growth and Capital Structure of Companies

731

(3) SZA. Firm size as measured by book value of total assets. The general
perception is that large firms are able to afford larger loans due to its large
asset base. Hence the relationship between leverage and size is expected to
be positive, and indeed many of the previous studies found a positive
relationship. Again, this may be true if firms indeed prefer debt to equity in
their financing hierarchy as in most developed countries. Positive
relationship is found in many of the previous studies. It would be interesting
to see if Asian managers manage capital structure in similar manner.
(4) SZS. Firm size as measured by total net sales. The expected relationship
for this variable is similar to the above variable, SZA.
(5) FAR. Fixed asset ratio is represented by the proportion of fixed asset to
total asset. Rajan and Zingales (1995) termed this ratio as tangibility.
Firms with high fixed asset component may be able to afford higher debt
because of higher collateral value of their assets. Hence the relationship is
expected to be positive.
(6) ROA. Profitability as measured by return on asset, that is, earnings before
interest and taxes over book value of total assets. The nature of
relationship of this variable against leverage is difficult to predict. For
firms that has reached their maturity stage, more internally generated
funds means less need for borrowing. On the other hand one may argue
that profitable firms may be easier to obtain a bank loan, which may result
in leverage increase, especially for growth firms.
(7) ROS. Profitability as measured by return on sales, that is, earnings before
interest and taxes over total sales. The expected relationship is similar to
the above variable, ROA.
(8) DOL. Degree of operating leverage is defined as the percentage in
earnings before interest and taxes divided by the percentage change in
sales. This variable measures the impact of fixed expenses used in a
company to enhance earnings. A high proportion of fixed costs means a
high business risk of the company, and lenders may be reluctant to
provide loan. Hence the expected relationship between this variable and
leverage is negative.
(9) DIV. Dividend policy is measured by annual gross dividend divided by
total earnings available for distribution. Since dividends are cash outflow
from the system, external funds need to be raised to finance firms
operation. If managers prefer debt to equity, a positive relationship may be
expected between this variable and leverage. But firms may pay high
dividends simply because there is no growth opportunity. In this situation
a non-positive relationship may appear. In most Asian countries in which
many industries are growing, we expect a positive relationship provided
managers prefer debt to equity in financing choice hierarchy.

Muhammad Mahmud

732

Regression analysis is employed on cross-sectional data from 1989 to 1998

inclusive. This research measures per capita GNP growth, prime lending rate,
creditors rights and enforcement of countrys economic development. Countrys
economic development may be represented by a number of indicators. Growth in per
capita is a barometer of economic activities. Interest rate is selected because it may
affect the demand for credit. The dummy variables are selected from the 1998-99
World Development Report to capture the specific financial liberalisation
phenomenon and countrys law and order situation. The following two measures are
represented by dummy variables:
(1) creditors rights;
(2) enforcement.
The other two measures use actual values:
(1) GNP growth per capita;
(2) prime lending rate.
5. RESULTS OF THE STUDY
5.1. Country Factor
Table 1 presents average leverage ratios included in this study by year and by
country. The first ratio is total liability to total asset (TLA). Although Japan and
Pakistan are at extreme ends of economic and capital market development, it is
somewhat surprising that they both show similar proportion of total liabilities to
asset ratio. Malaysia shows the lowest total liabilities ratio. Similar pattern is also
observed with the other two leverage ratios across the three countries.
One explanation for Japanese companies to be aggressive in their financing
structure is the Kieretsu system practiced in Japan, where the banking system has a
close relationship with the firms. Further, capital market in Japan is the most advance
among the three countries studied. Therefore it is not really surprising that Japanese
companies are highly geared. Rajan and Zingales (1995) have also documented this
evidence. But for Pakistan to have as high gearing as Japan is hard to explain. Since
its capital market is largely undeveloped, it may be the case that companies find it
difficult and expensive to float shares. The logical alternative for financing beyond
internal funds is debt, although interest rate in Pakistan is high compared to the other
countries (see also Table 2). Hence the high leverage of Pakistani companies is
dictated by the circumstances of the financial environment as opposed to by
managers free choice.
For Malaysia, firms are traditionally conservative in their capital structure
management. This has been documented in Kester and Isa (1994) and Tho (1993). It
is quite common to find many companies not to have any long-term debt in their

Table 1

1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
Average

Total Liabilities to Assets Ratio

(TLA)
Japan
Malaysia
Pakistan
N=505
N=109
N=104
.61
.43
.60
.61
.40
.66
.60
.39
.67
.60
.38
.66
.59
.40
.69
.59
.40
.67
.92
.40
.68
.92
.39
.69
.91
.40
.68
.90
.41
.70
.73
.40
.67

Average Leverage Ratios

Total Debt to Equity Ratio
(TDE)
Japan
Malaysia
Pakistan
N=505
N=109
N=104
2.44
1.30
2.35
2.45
1.02
2.88
2.36
0.91
2.97
2.38
0.80
3.02
2.34
0.90
3.40
2.29
0.92
3.03
2.71
0.92
3.21
2.69
0.91
3.02
2.70
0.89
3.09
2.71
0.90
3.18
2.51
0.95
3.02

Long-term Debt to Capital Ratio

(LTDC)
Japan
Malaysia
Pakistan
N=505
N=109
N=104
.35
.15
.24
.35
.13
.34
.36
.11
.35
.36
.11
.36
.36
.13
.35
.35
.14
.33
.34
.14
.32
.36
.14
.35
.35
.15
.32
.35
.14
.36
.35
.13
.33

Note: Total liabilities include all non-equity liabilities. Total debt includes long-term and short-term debts. All values are in book-value terms.

Table 2
Country Analysis: Comparative Facts and Statistics That May Be Relevant to Capital Structure Decision
Japan
Malaysia
Pakistan
Average TLA
0.64
0.40
0.64
Average TDE
2.55
0.98
2.83
Average LTDC
0.35
0.13
0.30
Accounting Standards
IASC & Japanese GAAP
IASC
IASC
Taxation

Corporate Tax Rate,

Prime Lending Rate
GNP Per Capita, 2001, USD
Market Capitalisation, (USD)*
GDP Growth****
Inflation** 2001
Savings % of GDP**
People per Telephone*
Literacy Rate***
*ASIAWEEK, June 23, 2000.
** ASIAWEEK, November 23, 2001.
***World Development Report 2003.

Interest Income (20%),

Dividend (20%, 35%),
Capital Gains (20%)
26-40 %
2.38 %
37950
3,795 Billion
0.6 %
0.8 %
30%
1.5
100%

Interest, Dividend Taxable

at Personal Rate, Capital
Gains Non Taxable
30%
7.75 %
3,531
159 Billion
0.4 %
1.4 %
47%
5.0
87%

Interest, Dividend Taxable

at 10%, Capital Gains Non
Taxable
30%-36%
11.53 %
450
10 Billion
2.7 %
3.4 %
15%
49.0
43%

Economic Growth and Capital Structure of Companies

735

capital structure. Their short-term debt and other liabilities are also at conservative
levels. It is hard to provide a rational explanation for this behaviour. One possible
reason is that it results from financing preferences of Malaysian business. As
documented by Kester and Isa (1994), the financing hierarchy of Malaysian
managers is quite different from those documented in the US. In Malaysia,
managers first choice of financing after internal funds is new equity as opposed to
debt. The packing order of financing for Malaysia is quite different from the
developed markets.
Table 1 also shows a remarkable stability in leverage ratios over the ten-year
period, 19891998 covered in this study. However there is a slight tendency for
Malaysia and Pakistan (more so for Pakistan) to show an increase in leverage over
the years. The tendency to increase leverage over time has long been observed for
the US market. This phenomenon should be expected because as the economy
develops, market environment becomes more competitive and the most efficient
form of financing would emerge.
Table 2 shows comparative facts and statistics on market and institutional
factors that may be relevant in determining capital structure in each of the three
countries. All the three countries employed an international standard for accounting
purposes. The tax environment, however, differs somewhat. Interest and dividend are
both taxable in all the three countries. However, capital gains are only taxable in
Japan, not in Malaysia and Pakistan. Malaysian corporate tax rate may be slightly
lower rate than Japan and Pakistan. But the gap may be too small to attribute to the
significant difference in the leverage structure discussed above.
Various market and economic indicators amply demonstrate the difference in
the stage of economic development between the three countries shown in Table 2. The
interest rate in 1998 is lowest in Japan and highest in Pakistan. And yet the leverage in
Pakistan is as high as Japan. As explained this may be due to the undeveloped equity
market. Underdeveloped capital market in Pakistan is reflected in the small market
capitalisation of the stock market and a very low GNP per capita, highest inflation and
lowest savings rate. The market capitalisation and GDP growth is below the optimal
level throughout much of the country and government spending is not always directed
towards the types of investment that have the most positive effects on growth.
5.2. Industry Factor
Many studies in the past have documented that there exists a significant
industry influence on capital structure. One of the reasons cited is that some
industries would require heavy investments in fixed assets, which has been found to
be a significant variable determining capital structure. This is because fixed assets
are closely related to firms collateral value and non-debt tax-shield. Other reason
cited is that some industries may have a higher cost of bankruptcy and financial
distress than other industries.

736

Muhammad Mahmud

Tables 3a to 3c respectively presents analyses of the three leverage ratios in

terms of industry influence. Each of the country samples is divided into 10 industry
sectors, and one-way ANOVA is performed to test if the mean industry ratios are
different from one another. The resultant F-statistics and associated probability
values are shown at the bottom of the table. The tables, however, show only six of
the ten industries analysed as these are the only common industry classification
between the three countries.
Tables 3a to 3c clearly show that the industry influence on the three leverage
ratios seems to be quite different from one country to another. For Japan, industry
factor is clearly significant for all leverage ratios. However, Malaysia and Pakistan
show significance for the liabilities ratio but marginal significance for the two debt
ratios. Across countries it is found that heavy industries, such as construction,
chemicals, electric and gas companies tend to have higher leverage ratios than other
industries. Light industries like food and beverages and transport services tend to
have low debt ratios. This is consistent with our prediction that the high proportion
of fixed assets required in these industries may drive leverage.
5.3. Size Factor
Firm size as measured by total asset is expected to have a positive influence on
leverage. This is because the larger the asset base of a company the more capable it is
to obtain a bank loan or to issue debt securities. Evidence on size effect on capital
structure has been somewhat mixed, but showed strong bias towards a positive
relationship. For example, Ferry and Jones (1979), Scott (1977), Chang and Rhee
(1990), Harris and Raviv (1991) and Homaifer and Benkato (1994) found positive
relationship between size and leverage. However, Titman and Wessel (1988), Fischer,
et al. (1989) finds that short-term debt is negatively related to leverage.
To study the effect of size on capital structure firms are divided into five size
groups. Firm size is based on year-end total assets, which are translated into US
dollar for common denomination. In this way firm size in each size group is the same
for all countries. The groups are revised annually. Tables 4a to 4c respectively
summarise the results of ANOVA analysis for each of the three ratios. Each table
shows the average means and its corresponding standard deviations for year 1989
and 1998, that is the beginning and ending of the period of study. F-statistics and
probability values are shown in the last two rows of the table.
For Japan, there is a clear demonstration of the size effect on capital structure
where large firms show a higher gearing compared to small firms. For total liabilities
ratio, the largest firm ratio is about 30 percent higher than that for the smallest firm.
For debt-equity ratio the difference is even greater, more than 100 percent, whereas
for long-term debt to capital the largest firm ratio is about 72 percent higher than the
smallest firm. Comparing the results for the year at the beginning of the study period,
1989 and the year at the end, 1998 it can be observed that there is a remarkable

Table 3a

Industry Type
Chemicals
(N=94,6,8)
Construction
(N=55,20,5)
Electric
Machinery
(N=97,5,5)
Food and
Beverages
(N=42,12,8)
Natural
Resources
(N=25,21,6)
Transport
Equipment
(N=50,6,7)
F-ratio
Prob-value

Total Liabilities to Assets Ratio by Industrial Sectors in the Three Countries

Japan
Malaysia
Pakistan
1989
1998
1989
1998
1989
1998
.625
.570
.356
.439
.659
.655
(.183 )
(174)
(.158 )
(.156 )
(.215 )
(.263 )
.797
.777
.387
.425
.664
.756
(.101 )
(.136 )
(.215 )
(.138 )
(.119 )
(.101 )
.546
(.197 )

.548
(.195 )

.411
(.168 )

.487
(.215 )

.747
(.055 )

.715
(.094 )

.567
(.148 )

.552
(.165 )

.425
(.166 )

.445
(.187 )

.509
(.187 )

.797
(.102 )

.769
(.144 )

.765
(.121 )

.207
(.172 )

.276
(.160 )

.626
(.233 )

.814
(.098 )

.663
(.161 )
14.15
0.001

.659
(.183 )
11.57
0.000

.344
(.151 )
3.26
0.002

.433
(.105 )
2.48
0.016

.673
(.207 )
2.84
0.005

.739
(.125 )
3.55
0.002

Notes: 1. The figures in parenthesis indicate standard deviation.

2. Many of the industrial sectors do not have similar labels across countriesthey are assumed similar based on broad classification. Listed below are
specific country classification of the groups that have been assumed similar:
Construction: Construction for Japan, Properties and Construction for Malaysia and Building Material for Pakistan.
Electric Machinery for Japan, Electricals for Malaysia and Cable and Electric for Pakistan.
Natural Resources: Electric, Power and Gas for Japan, Plantation for Malaysia, and Gas and Energy for Pakistan.

Table 3b

Industry Type
Chemicals
(N=94,6,8)
Construction
(N=55,20,5)
Electric
Machinery
(N=97,5,5)
Food and
Beverages
(N=42,12,8)
Natural
Resources
(N=25,21,6)
Transport
Equipment
(N=50,6,7)
F-ratio
Prob-value

Total Debt to Equity Ratio by Industrial Sectors in the Three Countries

Japan
Malaysia
Pakistan
1989
1998
1989
1998
1989
1998
2.77
2.09
.643
.904
4.06
3.65
(2.83 )
(2.62 )
(.466 )
(.559 )
(.5 .32)
(.322 )
5.44
5.55
.853
.832
2.21
3.08
(3.66 )
(4.02 )
(.757 )
(.431 )
(.852 )
(.132 )
1.99
(2.22 )

1.76
(1.47 )

.818
(.513 )

1.45
(1.18 )

2.96
(.812 )

2.89
(1.66 )

1.64
(.97 )

1.61
(1.16 )

.915
(.666 )

1.15
(.866 )

1.46
(.139 )

6.14
(5.87 )

5.11
(4.17 )

4.15
(2.20 )

.349
(.472 )

.458
(.384 )

3.16
(3.13 )

5.95
(3.86 )

3.12
(2.90 )
12.29
0.000

2.95
(2.39 )
12.46
0.000

.583
(.345 )
4.03
0.003

.855
(.257 )
2.12
0.038

3.59
(3.81 )
1.26
0.267

4.09
(3.76 )
1.92
0.066

Notes: 1. The figures in parenthesis indicate standard deviation.

2. Many of the industrial sectors do not have similar labels across countriesthey are assumed similar based on broad classification. Listed below are
specific country classification of the groups that have been assumed similar:
Construction: Construction for Japan, Properties and Construction for Malaysia and Building Material for Pakistan.
Electric Machinery for Japan, Electricals for Malaysia and Cable and Electric for Pakistan.
Natural Resources: Electric, Power and Gas for Japan, Plantation for Malaysia, and Gas and Energy for Pakistan.

Table 3c

Industry Type
Chemicals
(N=94,6,8)
Construction
(N=55,20,5)
Electric
Machinery
(N=97,5,5)
Food and
Beverages
(N=42,12,8)
Natural
Resources
(N=25,21,6)
Transport
Equipment
(N=50,6,7)
F-ratio
Prob-value

Long-term Debt to Capital Ratio by Industrial Sectors in the Three Countries

Japan
Malaysia
Pakistan
1989
1998
1989
1998
1989
1998
.376
.312
.082
.189
.169
.354
(.198)
(.199)
(.098)
(.156)
(.178)
(.103)
.352
.398
.133
.191
.335
.546
(.142)
(.207)
(.132)
(.143)
(.289)
(.175)
.296
(.176)

.276
(.166)

.076
(.042)

.258
(.253)

.187
(.146)

.278
(.495)

.274
(.154)
.618
(.223)

.280
(.169)

.104
(.094)

.090
(.084)

.071
(.119)

.109
(.047)

.624
(.228)

.064
(.085)

.086
(.115)

.296
(.265)

.318
(.395)

.360
(.196)
14.26
0.000

.351
(.190)
11.21
0.000

.093
(.090)
1.34
0.249

.108
(.143)
1.89
0.064

.331
(.253)
1.95
0.054

.344
(.193)
1.84
0.085

Notes: 1. The figures in parenthesis indicate standard deviation.

2. Many of the industrial sectors do not have similar labels across countriesthey are assumed similar based on broad classification. Listed below are
specific country classification of the groups that have been assumed similar:
Construction: Construction for Japan, Properties and Construction for Malaysia and Building Material for Pakistan.
Electric Machinery for Japan, Electricals for Malaysia and Cable and Electric for Pakistan.
Natural Resources: Electric, Power and Gas for Japan, Plantation for Malaysia, and Gas and Energy for Pakistan.

740

Muhammad Mahmud

Table 4a
Analysis of Variance for Mean Differences among Five Size Groups in
Terms of Total Liabilities to Total Assets Ratio
Japan
Malaysia
Pakistan
Size Quintile
1989
1998
1989
1998
1989
1998
Group 1
.6127
.5569
.3050
.3050
.5887
.8609
(Smallest)
(.195 ) (.196 ) (.215 ) (.185 ) (.181 ) (.245 )
Group 2
.6237
.5734
.4130
.4290
.5058
.5906
(.179 ) (.187 ) (.207 ) (.250 ) (.206 ) (.249 )
Group 3
.6132
.6063
3590
.3960
.6243
.7427
(.192 ) (.200 ) (.158 ) (.118 ) (.179 ) (.076 )
Group 4
.7232
.6460
.4040
.3790
.6456
.6177
(.170 ) (.183 ) (.240 ) (.170 ) (.173 ) (.206 )
Group 5
.7412
.7170
.4820
.4300
.7089
.6780
(Largest)
(.153 ) (.165 ) (.153 ) (.187 ) (.163 ) (.163 )
F-ratio
9.166
12.72
1.968
1.009
3.043
1.765
Prob-value
0.000
0.000
0.104
0.409
0.021
0.144
Note: Figures in parentheses are standard deviations.

Table 4b
Analysis of Variance for Mean Differences among the Five Size
Groups in Terms of Debt-Equity Ratio
Japan
Malaysia
Pakistan
Size Quintile
1989
1998
1989
1998
1989
1998
Group 1
2.470
1.890
.6130
.5550
1.860
4.430
(Smallest)
(2.33)
(1.74)
(.614)
(.530)
(1.11)
(7.14)
Group 2
2.760
1.860
.9840
1.259
1.730
2.010
(3.20)
(1.41)
(.909)
(1.44)
(2.49)
(1.27)
Group 3
2.560
2.510
.6730
.7040
2.680
3.230
(2.45)
(2.66)
(.627)
(.285)
(2.78)
(1.57)
Group 4
4.400
3.380
1.110
.7440
2.900
3.020
(4.03)
(3.98)
(1.30)
(.525)
(3.51)
(4.15)
Group 5
4.580
4.010
1.100
1.046
4.420
3.260
(Largest)
(3.71)
(3.21)
(.687)
(1.00)
(4.68)
(3.20)
F-ratio
9.278
13.07
1.571
1.482
1.958
0.316
Prob-value
0.000
0.000
0.188
0.213
0.106
0.865
Note: Figures in parentheses are standard deviations.

Economic Growth and Capital Structure of Companies

741

Table 4c
Analysis of Variance for Mean Differences among the Five Size
Groups in Terms of Long-term Debt to Capital
Japan
Malaysia
Pakistan
Size Quintile
1989
1998
1989
1998
1989
1998
Group 1
.311
.267
.060
.106
.274
.086
(Smallest)
(.188 )
(.182 )
(.070 )
(.138 )
(.244 )
(.220 )
Group 2
.338
.289
.112
.128
.138
.219
(.170 )
(.173 )
(.112 )
(.160 )
(.195 )
(.222 )
Group 3
.345
.325
.137
.110
.330
.275
(.167 )
(.180 )
(.126 )
(.120 )
(.286 )
(.219 )
Group 4
.449
.377
.163
.113
.225
.255
(.180 )
(.205 )
(.182 )
(.123 )
(.214 )
(.221 )
Group 5
.524
.461
.209
.168
.310
.382
(Largest)
(.220 )
(.228 )
(.137 )
(.152 )
(.217 )
(.212 )
F-ratio
18.90
16.96
3.89
0.95
2.19
2.90
Prob-value
0.000
0.000
0.005
0.437
0.076
0.026
Note: Figures in parentheses are standard deviations.

stability in the size effect. The results for Japan are consistent with most of the
previous studies on other developed markets.
The results for both Malaysia and Pakistan do not seem to indicate strong
presence of the size effect in capital structure. For Malaysia although the smallest
firm group tend to show the lowest gearing and the largest firm shows the highest,
but the relationship of intermediate size groups is not monotonous. For Pakistan
there is no clear trend that can be observed. However, the second smallest group
seems to show the lowest leverage ratios. Interestingly, the highest liabilities ratio
and debt-equity ratio shifted from the largest firm group in the beginning of the
period to the smallest group firm at the end of the period. This trend, however, is not
observed for the long-term debt to capital ratio. In summary, the results in this
section indicate a clear presence of capital structure size effect in Japan, but not in
Malaysia and Pakistan.
5.4. Firm-specific Factors
If firms make capital structure decisions as if there exist an optimal mix
between debt and equity, it would be both interesting and useful to know what are
the factors that determine such decisions. Many researchers, using many
firm-specific variables, have studied this area of investigation. Based on a survey on
previous studies we came up with more than a dozen variables. However, some of
them are simply variations of each other. After performing a correlation test, we

Muhammad Mahmud

742

finally arrive at nine firm-specific variables for the current study. A pooled
time-series cross sectional regression was run with the leverage ratios as dependent
variables against the nine firm-specific factors as independent variables. The
following three regressions are run for each country:
TLA; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR; + 6ROA; +
7ROS; + 8DOL; + 9DIV;

(1)

TDE; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR; + 6ROA; +

7ROS; + 8DOL; + 9DIV;

(2)

LTDC; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR; + 6ROA; +

7ROS; + 8DOL; + 9DIV;

(3)
Where:
TLA =
TDE =
LTDC =
GRA =
GRS =
SZA =
SZS =
FAR =
ROA =
ROS =
DOL =
DIV =

total liabilities to total asset ratio.

total debt to equity ratio.
long-term debt to capital ratio.
annual growth in asset.
annual growth in sales.
total asset.
total sales.
fixed asset ratio = net fixed assets = total assets.
annual return on asset = EBIT = total asset.
return on sales = EBIT = sales.
degree of operating leverage = A percent in EBIT = o percent in sales.
dividend payout ratio = total gross dividend paid = total earnings after
tax.

The results of the regressions are presented in Tables 5a to 5c. Looking across
the tables the results indicates that each leverage ratio is driven by different factors,
and the factors are also different across countries. This makes it difficult to draw
general observations across debt ratios and across countries.
For total liabilities ratio, total assets, sales, fixed asset ratio, return on assets
and return on sales, but some with unexpected signs drive Japan. The negative
relationship between liabilities and sales goes against our prediction. Similarly, the
relationship is also negative for return on sales. One possible explanation is that as
sales increases, and as profitability increases, more internal funds are generated
and this leads to a lesser need for external financing. In fact the internal funds thus
generated may be used to retire debt. This is especially true for mature industries
where there is zero or little growth as may be the case with many Japanese
industries. The negative relationship with the fixed asset ratio is difficult to
explain.

Economic Growth and Capital Structure of Companies

743

Table 5a
Regression Results of Total Liabilities to Total Assets (TLA)
Ratio on the Firm-specific Factors
TLA; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR;
+ 6ROA; + 7ROS; + 8DOL; + 9DIV;
Japan (N=505)
Coefficient
t-stat

Variables

Malaysia (N=109)
Coefficient
t-stat

Pakistan (N=104)
Coefficient
t-stat

Constant

0.5662

40.5

0.4983

13.52

0.6537

16.15

GRA: Growth in Assets

0.0445

1.49

0.1851

3.21**

0.3545

5.49**

GRS: Growth in Sales

0.0276

1.25

0.2587

5.35**

0.0396

0.74

SZA: Total Assets

0.0234

6.22**

0.0547

2.14*

0.0542

0.57

SZS: Net Sales

0.0263

4.52**

0.0296

1.02

0.1646

2.29*

FAR: Fixed Asset Ratio

0.1001

5.63**

0.1975

5.03**

0.0655

1.30

ROA: Return on Assets

0.1348

3.01**

0.7316

6.35**

0.4003

6.58**

ROS: Return on Sales

0.3945

10.01**

0.0131

1.06

0.2243

1.22

0.0002

0.15

0.0005

1.22

0.2336

1.25

0.007

1.31

0.0422

1.60

0.0135

1.99

DOL: Degree of Operating

Leverage
Div: Dividend Payout Ratio
Adj. R2

0.83

0.56

0.79

*Significant at least at 5 percent level.

**Significant at least at 1 percent level.

Table 5b
Regression Results of Total Debt to Equity (TDE)
Ratio on the Firm-specific Factors
TDE; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR;
+ 6ROA; + 7ROS; + 8DOL; + 9DIV;
Variables

Japan (N=505)
Coefficient
t-stat

Malaysia (N=109)
Coefficient
t-stat

Pakistan (N=104)
Coefficient
t-stat

Constant

3.3313

17.01

1.4724

9.96

3.3461

GRA: Growth in Assets

1.8731

4.17**

0.5892

1.82

2.0841

4.88
1.86

GRS: Growth in Sales

0.3323

0.99

0.8477

3.93**

0.8981

0.85

SZA: Total Assets

0.5904

6.59**

0.0620

0.48

3.5546

2.21*

SZS: Net Sales

0.5401

5.55**

0.1263

0.94

0.5153

3.97**

FAR: Fixed Asset Ratio

1.3908

4.52**

0.7181

4.44**

0.8189

0.85

ROA: Return on Assets

0.3301

0.42

4.2126

7.21**

0.7160

6.44**

ROS: Return on Sales

2.9851

4.94**

0.0891

1.82

1.5090

3.01**

0.0003

0.22

0.0018

0.78

0.0012

3.16**

Ratio

0.00084

1.12

0.1939

0.15

0.4215

1.88

Adj. R2

0.77

DOL: Degree of Operating

Leverage
Div: Dividend Payout

*Significant at least at 5 percent level.

**Significant at least at 1 percent level.

0.27

0.68

744

Muhammad Mahmud

Table 5c
Regression Results of Long-term Debt to Capital (LTDC)
Ratio on Firm-specific Factors
LTDC; = ; + 1GRA; + 2GRS;+ 3SZA; + 4SZS; + 5FAR;
+ 6ROA; + 7ROS; + 8DOL; + 9DIV;
Japan (N=505)
Variables
Coefficient
t-stat
Constant
0.3280
22.44
GRA: Growth in
Assets
0.1547
5.22**
GRS: Growth in
Sales
0.1514
5.42**
SZA: Total
Assets
0.0395
5.54**
SZS: Net Sales
0.0510
1.88
FAR: Fixed Asset
Ratio
0.1022
5.90**
ROA: Return on
Assets
0.0408
0.70
ROS: Return on
Sales
0.2724
5.38*
DOL: Degree of
Operating
0.0001
0.09
Leverage
Div: Dividend
Payout Ratio
0.0007
1.41
0.76
Adj. R2
*Significant at least at 5 percent level.
**Significant at least at 1 percent level.

Malaysia (N=109)
Coefficient
t-stat
0.1513
5.25

Pakistan (N=104)
Coefficient
t-stat
0.0651
1.63

0.2302

5.66**

0.2448

3.26**

0.0951

2.78**

0.0872

1.31

0.0564
0.0434

3.10**
2.23**

0.4238
0.1240

3.80**
1.55

0.0137

0.70

0.3574

5.48**

0.4344

5.47**

0.4890

6.42**

0.0125

1.87

0.1581

0.77

0.0003

0.90

0.2408

0.87

0.0014
0.44

0.09

0.0224
0.73

2.42*

For Malaysia, the liabilities ratio is driven by growth in assets and sales, total
assets, fixed asset ratio and return on sales. For Pakistan the driving factors are
growth in assets, sales and return on assets. The only common variable across
countries is return on assets. However, the relationship for Japan is positive whereas
it is negative for Malaysia and Pakistan. The positive relationship is consistent with
our expectation and with our earlier results on the existence of the size effect for
Japan. The negative for Malaysia and Pakistan may be superior as our earlier
analysis on size effect show no significant relationship.
Table 5b on total debt to equity ratio shows similar confusion as those in
Table 5a. However, Table 5c on long-term debt to capital ratio shows an interesting
result, that is, two variables are found to be important across countries in explaining
long-term debt ratio. The variables are growth in assets and total assets, and both
show a positive relationship with long-term debt ratio. This result is consistent with
our expectation.

Economic Growth and Capital Structure of Companies

745

For Japan, it is found that there are two pervasive variables that show
significant relationship with all three leverage ratios. These are total assets, with a
positive sign and return on sales with a negative sign. One possible interpretation is
that profitable firms with large asset base are using their internally generated funds to
reduce debt burden.
There are also two pervasive variables for Malaysia, but the variables are
different from those for Japan. Malaysias leverage is positively related to growth in
sales, but negatively related to return on assets. The positive relationship between
leverage and growth in sales is consistent with the findings of previous researchers.
It can be explained by the fact that growth companies are successful companies and
may be easier to obtain loans. But the negative relationship between leverage and
return on assets seem to indicate an opposite effect, that is profitable and efficiently
run firms tend to reduce their debt burden. Similar explanation can be offered to the
situation in Pakistan where only one pervasive variable is found, that is return on
asset, which is negatively related with leverage.
6. CONCLUSION
The current study investigates factors affecting firms capital structure
decisions in three Asian countries: Japan, Malaysia and Pakistan. These countries are
chosen to represent different stages of economic and corporate environments. The
factors considered are capital market development, industry and firm factors. Firm
factors include growth, size, fixed asset ratio, profitability, operating leverage and
dividend policy.
It is found that firms in Japan, and surprisingly in Pakistan show very high
leverage ratios with total debt to capital ratio amounting to more than 70 percent. For
Malaysia the ratio is about 50 percent. The high gearing for Japanese companies is to
be expected in view of its developed market status. But for Pakistan, the gearing is
more due to undeveloped capital market which forces firms to opt for bank loans as
opposed to raising new equities. Good economic policy requires both increasing the
market capitalisation and reorienting of government spending from consumption to
investment in physical capital stock. Malaysias conservative financing management
may be due to the lack of competition in the market.
As a powerful anti-poverty tool, micro-credit has demonstrated relevance to
poor people. Micro credit programmes extend small loans to poor people for selfemployment projects that generate income, allowing them to care for themselves and
their families. In most cases, micro credit programmes offer a combination of
services and resources to their clients in addition to credit for self-employment.
These often include savings, training, network and peer support. Interest rates are
high in Pakistan as compared to Japan and Malaysia. The Government of Pakistan
should undertake the following reforms.

746

Muhammad Mahmud

Strengthen the legal and judicial reform laws to allow financial institutions
to foreclose on collateral in the case of unpaid loans without going through
lengthy court proceedings.
Improve the National Savings Scheme.
Allow and encourage consolidation of small financial institutions to reduce
fragmentation in the financial sector.
Industry influence is the strongest in Japan and Pakistan. Textile industry is
the largest industry in Pakistan. It has an important role in the development of the
economy (about 9 percent share in GDP). The study showed that textile industry is
very much neglected in Pakistan. The Government should take such measures that
could increase the quality production of cloth and export of textile products to other
countries.
For Malaysia, there are some elements of industry influence, but not
consistent across different leverage ratios. Lack of industry influence in Malaysia
may be due to aggressive diversification of activities at firm level.
The results for firm-specific factors are largely mixed. Size factor is positively
related to leverage in Japan, but not in the other two countries studied. There is a
tendency for asset related variables, such as growth in asset, total asset, fixed asset ratio
and return on asset to be important determinant of capital structure, while sales related
variables such as sales growth, total sales and return on sales to be less important.
Degree of operating leverage and dividend policy are found to be not important.
It was noted that growth in GNP per capita, which proxies for economic
activities in the country, was significantly affecting growth in the capital structure of
companies in Japan and Malaysia. Although this variable was insignificant when
regressed inclusive three countries. For Pakistan this variable remains insignificant
with all the three leverage ratios. The interest rate, which is measured, by prime
lending rate, is a major decisive factor affecting demand for credit in Japan and
Malaysia. Japanese companies are more leveraged companies than Malaysian
companies. Therefore Japanese companies save tax and take more debt. Malaysian
companies are risk averse and look consciously at interest. Prime lending rate
appeared unexpectedly positively related to three leverage ratios in Pakistan.
Pakistani companies take risk therefore very low investment is carried out in the
manufacturing sector.
It is revealed from the analysis that creditors rights provide significant impact
on capital structure of companies and overall corporate sector in all the three
countries. The indicator law enforcement appeared negatively significant with debt
to equity and long-term debt to total capital ratio. This indicator was, perhaps not
representing the countrys overall law and order situation in Japan, Malaysia, and
Pakistan.

Economic Growth and Capital Structure of Companies

747

It was noted that growth in GNP per capita, which proxies for economic
activities in the country, was significantly affecting growth in the capital structure of
companies in Japan and Malaysia. Although this variable was insignificant when
regressed inclusive three countries. For Pakistan this variable remains insignificant
with all the three leverage ratios. The interest rate, which is measured, by prime
lending rate, is a major decisive factor affecting demand for credit in Japan and
Malaysia. Japanese companies are more leveraged companies than Malaysian
companies. Therefore Japanese companies save tax and take more debt. Malaysian
companies are risk averse and look consciously at interest. Prime lending rate
appeared unexpectedly positively related to three leverage ratios in Pakistan.
Pakistani companies take risk therefore very low investment is carried out in the
manufacturing sector.
It is revealed from the analysis that creditors rights provide significant impact
on capital structure of companies and overall corporate sector in all the three
countries. The indicator law enforcement appeared negatively significant with debt
to equity and long-term debt to total capital ratio. This indicator was, perhaps not
representing the countrys overall law and order situation in Japan, Malaysia, and
Pakistan.
REFERENCES
Allen, D. E., and H. Mizuno (1989) The Determinants of Corporate Capital
Structure: Japanese Evidence. Applied Economics 21, 569585.
Baskin, J. (1989) An Empirical Investigation of the Pecking Order Hypothesis.
Financial Management 2635.
Chang, R. P., and S. G. Rhee (1990) The Impact of Personal Taxes on Corporate
Dividend Policy and Capital Structure Decisions. Financial Management 2131.
Ferry, M. G., and W. H. Jones (1979) Determinants of Financial Structures: A New
Methodological Approach. The Journal of Finance 34:3, 631644.
Fischer, E. O., R. Heinkel, and J. Zechner (1989) Dynamic Capital Structure Choice:
Theory and Test. The Journal of Finance 44, 1940.
Graham, Barry, and Garry Bromson (1992) Industry and Country Influences on
Capital Structure: An Empirical Study of Six Asian-Pacific Countries. Asian
Review of Accounting 1:2, 102119.
Harris, M., and A. Raviv (1991) The Theory of Capital Structure. The Journal of
Finance 297351.
Homaifar, G., J. Zietz, and O. Benkato (1994) An Empirical Model of Capital
Structure: Some New Evidence. Journal of Business Finance and Accounting
21:1, 114.
Johnson, S. A. (1997) An Empirical Analysis of the Determinants of Corporate Debt
Ownership Structure. Journal of Financial and Quantitative Analysis 32:1, 47
69.

748

Muhammad Mahmud

Miller, M. H. (1997) Debt and Taxes. The Journal of Finance 32:2, 261275.
Modigliani, F., and M. H. Miller (1958) The Cost of Capital, Corporate Finance and
the Theory of Investment. The American Economic Review 48:3, 261297.
Modigliani, F., and M. H. Miller (1963) Corporate Income Taxes and the Cost of
Capital: A Correction. The American Economic Review 433442.
Nakamura, A., and M. Nakamura (1982) On the Firms Production, Capital Structure
and Demand for Debt. The Review of Economics and Statistics 64, 384393.
Rajan, R. G., and L. Zingales (1995) What Do We Know About Capital Structure?
Some Evidence From International Data. The Journal of Finance 50:5, 1421
1460.
Remmers, L., A. Stonehill, R. Wright, and T. Beckhuisen (1974) Industry and Size
as Debt Ratio Determinants in Manufacturing Internationally. Financial
Management 2432.
Scott, D. F., and J. D. Martin (1975) Industry Influence of Financial Structure.
Financial Management 6773.
Scott, J. H. Jr. (1977) Bankruptcy, Secured Debt, and Optimal Capital Structure. The
Journal of Finance 32:1, 119.
Tho Lai Mooi (1993) Financial Structure Strategies: Some Evidence from Malaysian
Listed Companies. Securities Industry Review 19:2, 6371.
Titman, S. (1984) The Effect of Capital Structure on a Firms Liquidating Decision.
The Journal of Financial Economics 13, 137151.
Titman, Sheridan, and Roberto Wessels (1988) The Determinants of Equity Risk.
The Journal of Finance 43, 119.
Toy, N., A. Stonehill, L. Remmers, R. Wright, and T. Beekhuisen (1974) A
Comparative International Study of Growth, Profitability, and Risk as
Determinants of Corporate Debt Ratios in the Manufacturing Sector. Journal of
Financial and Quantitative Analysis 87588.